一种选择性加权聚类融合算法  

Selective and weighted clustering fusion algorithm

作　　者：樊晓平[1] 盛荣芬[1] 廖志芳[1] 刘丽敏[1] 

机构地区：[1]中南大学信息科学与工程学院,长沙410075

出　　处：《计算机工程与应用》2012年第22期195-200,共6页Computer Engineering and Applications

基　　金：湖南省自然科学基金(No.06JJ50143)

摘　　要：传统的聚类融合方法通过融合所有成员实现融合,无法彻底消除劣质聚类成员对融合质量的影响,而从聚类成员的选择和加权两方面进行聚类融合,即先采用两两融合技术代替融合所有聚类结果进行聚类成员选择,然后进行基于属性的聚类成员加权,在理论上具有更好优越性。通过对真实数据和模拟数据的实验发现,该算法能有效处理聚类成员的质量差异,比传统聚类融合能得到更好的聚类结果,具有较好可扩展性。Traditional clustering fusion method is to integrate all obtained cluster member,;, it doesn＇t eliminate the inferior quality＇ influence in the integration completely, but it can have better advantages in theory from the selec- tion and weight of clustering members to make clustering fusion, which select cluster members through fusioning any two clustering members instead of all clustering members once, and weight cluster m~：mbers through the meth- od of attribute weighting. It does experiment in the real and simulated datas. The result shows that the clustering al- gorithm can effectively deal with the difference of members quality, compared to the traditional clustering fusion method, it can get better results and have a good expansibility.

关 键 词：聚类融合 聚类成员选择 聚类成员加权 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]